mrpp {vegan} | R Documentation |
Multiple Response Permutation Procedure (MRPP) provides a test of whether there is a significant difference between two or more groups of sampling units.
mrpp(dat, grouping, permutations = 999, distance = "euclidean", weight.type = 1, strata) meandist(dist, grouping, ...) ## S3 method for class 'meandist': summary(object, ...) ## S3 method for class 'meandist': plot(x, cluster = "average", ...)
dat |
data matrix or data frame in which rows are samples and columns are response variable(s), or a dissimilarity object or a symmetric square matrix of dissimilarities. |
grouping |
Factor or numeric index for grouping observations. |
permutations |
Number of permutations to assess the significance of the MRPP statistic, delta. |
distance |
Choice of distance metric that measures the
dissimilarity between two observations . See vegdist for
options. This will be used if dat was not a dissimilarity
structure of a symmetric square matrix. |
weight.type |
choice of group weights. See Details below for options. |
strata |
An integer vector or factor specifying the strata for permutation. If supplied, observations are permuted only within the specified strata. |
dist |
A dist object of dissimilarities, such as
produced by functions dist , vegdist or
designdist . |
object, x |
A meandist result object. |
cluster |
A clustering method for the hclust
function. Any hclust method can be used, but perhaps only
"average" and "single" make sense. |
... |
Further arguments passed to functions. |
Multiple Response Permutation Procedure (MRPP) provides a test
of whether there is a significant difference between two or more groups
of sampling units. This difference may be one of location (differences
in mean) or one of spread (differences in within-group
distance). Function mrpp
operates on a data.frame
matrix
where rows are observations and responses data matrix. The response(s)
may be uni- or multivariate. The method is philosophically and
mathematically allied with analysis of variance, in that it compares
dissimilarities within and among groups. If two groups of sampling units
are really different (e.g. in their species composition), then average
of the within-group compositional dissimilarities ought to be less than
the average of the dissimilarities between two random collection of
sampling units drawn from the entire population.
The mrpp statistic delta is simply the overall weighted mean of
within-group means of the pairwise dissimilarities among sampling
units. The correct choice of group weights is currently not clear. The
mrpp
function offers three choices: (1) group size (n), (2) a
degrees-of-freedom analogue (n-1), and (3) a weight that is the number
of unique distances calculated among n sampling units (n(n-1)/2).
The mrpp
algorithm first calculates all pairwise distances in the
entire dataset, then calculates delta. It then permutes the
sampling units and their associated pairwise distances, and recalculates
a delta based on the permuted data. It repeats the permutation
step permutations
times. The significance test is simply the
fraction of permuted deltas that are less than the observed delta, with
a small sample correction. The function also calculates the
change-corrected within-group agreement
A = 1 -delta/E(delta), where E(delta) is the expected
delta assessed as the average of permutations.
With weight.type = 3
, the function also calculates classification
strength (Van Sickle 1997) which is defined as the difference between
average between group dissimilarities and within group
dissimilarities. With weight.type = 3
the classification strength
is a simple transformation of delta, and has the same permutation
significance.
If the first argument dat
can be interpreted as dissimilarities,
they will be used directly. In other cases the function treats
dat
as observations, and uses vegdist
to find
the dissimilarities. The default distance
is Euclidean as in the
traditional use of the method, but other dissimilarities in
vegdist
also are available.
Function meandist
calculates a matrix of mean within-cluster
dissimilarities (diagonal) and between-cluster dissimilarites
(off-diagonal elements), and an attribute n
of grouping
counts
. Function summary
finds the within-class, between-class
and overall means of these dissimilarities, and the MRPP statistics
with all weight.type
options and the classification
strength. The function does not allow significance tests for these
statistics, but you must use mrpp
with appropriate
weight.type
. Function plot
draws a dendrogram of the
result matrix with given cluster
method (see
hclust
). The terminal segments hang to within-cluster
dissimilarity. If some of the clusters is more heterogeneous than the
combined class, the leaf segment is reversed.
The function returns a list of class mrpp with following items:
call |
Function call. |
delta |
The overall weighted mean of group mean distances. |
E.delta |
expected delta, under the null hypothesis of no group structure. This is the mean of original dissimilarities. |
CS |
Classification strength (Van Sickle 1997) with
weight.type = 3 and NA with other weights. |
n |
Number of observations in each class. |
classdelta |
Mean dissimilarities within classes. The overall
delta is the weighted average of these values with given
weight.type |
Pvalue |
Significance of the test. |
A |
A chance-corrected estimate of the proportion of the distances explained by group identity; a value analogous to a coefficient of determination in a linear model. |
distance |
Choice of distance metric used; the "method" entry of the dist object. |
weight.type |
The choice of group weights used. |
boot.deltas |
The vector of "permuted deltas," the deltas calculated from each of the permuted datasets. |
permutations |
The number of permutations used. |
This difference may be one of location (differences in mean) or one of
spread (differences in within-group distance). That is, it may find a
significant difference between two groups simply because one of those
groups has a greater dissimilarities among its sampling units. Most
mrpp
models can be analysed with adonis
which seems
not suffer from the same problems as mrpp
and is a more robust
alternative.
M. Herny H. Stevens HStevens@muohio.edu and Jari Oksanen.
B. McCune and J. B. Grace. 2002. Analysis of Ecological Communities. MjM Software Design, Gleneden Beach, Oregon, USA.
P. W. Mielke and K. J. Berry. 2001. Permutation Methods: A Distance Function Approach. Springer Series in Statistics. Springer.
J. Van Sickle 1997. Using mean similarity dendrograms to evaluate classifications. Journal of Agricultural, Biological, and Environmental Statistics 2:370-388.
anosim
for a similar test based on ranks, and
mantel
for comparing dissimilarities against continuous
variables, and
vegdist
for obtaining dissimilarities,
adonis
is a more robust alternative in most cases.
data(dune) data(dune.env) dune.mrpp <- mrpp(dune, dune.env$Management) dune.mrpp # Save and change plotting parameters def.par <- par(no.readonly = TRUE) layout(matrix(1:2,nr=1)) plot(dune.ord <- metaMDS(dune), type="text", display="sites" ) ordihull(dune.ord, dune.env$Management) with(dune.mrpp, { fig.dist <- hist(boot.deltas, xlim=range(c(delta,boot.deltas)), main="Test of Differences Among Groups") abline(v=delta); text(delta, 2*mean(fig.dist$counts), adj = -0.5, expression(bold(delta)), cex=1.5 ) } ) par(def.par) ## meandist dune.md <- meandist(vegdist(dune), dune.env$Management) dune.md summary(dune.md) plot(dune.md)